|
 The water "bug" that ravaged Milwaukee proved to be a micro-needle in a public haystackLaura Conrad
In 1993, a major outbreak of a water borne illness attributed to exposure to Cryptosporidium was reported in Milwaukee, WI. More than 100 people died and about 400,000 became dehydrated and developed intestinal problems ranging from mild to severe. At the time, finding the culprit was a serious challenge for public health officials. Cryptosporidium is a protozoan that can sometimes be found in drinking water even after the water has been treated to make it presumably potable. Another protozoan that is often found in rivers and can persist in drinking water is Giardia. At the time of the Milwaukee crisis, Cryptosporidium was a microbe we knew little about. Where is it found, where did it come from, what are its occurrence and sources, what is the health risk, does it need to be regulated, and if so, how will it be monitored - all were questions that had to be answered by public health officials and lawmakers. It is notable that the Milwaukee outbreak occurred at about the same time the Safe Drinking Water Act (SDWA) was up for reauthorization (see Today's Chemist at Work February 1997 for a discussion of the SDWA). Many people believe that the Milwaukee Cryptosporidium episode may well have been the impetus needed to move Congress to begin serious deliberations that led to the eventual renewal of the SDWA. Responding to public concerns, Congress passed a new version of the act in 1996 that required the US Environmental Protection Agency (EPA) to replace a longstanding priority on chemical pollutant regulation with a policy that at least balances chemical and biological hazard regulation of drinking water. The problem is that we know a great deal more about chemical contamination of drinking water than about biological contamination. Chemicals are relatively easy to identify and quantify. We have chromatographs to separate one species from another and spectrometers to identify them. That is not the case with biological contamination. Many biological methods are basically visual, requiring a specially trained biologist to grow a colony of bacteria for inspection or to look through a microscope to identify the protozoa. One might at first think that the solu tion to microbiological contamination would be to treat water with disinfection chemicals such as chlorine or ozone, irrespective of whether the microbes are present. The problem is that the cure can cause more problems than the microbe it potentially removes. All drinking water contains small quantities of humic acid, a complex organic species resulting from the natural degradation of plants in rivers, streams, and lakes. Humic acid alone is not harmful in the trace quantities normally present in finished or potable water. However, when treated with chlorine, ozone, or similar disinfection agents, humic acid can break down to form bromate (BrO3-), chlorate (ClO3-), and chlorite (ClO2-) ions, all known to be human toxins when ingested in large quantities. These chemicals are known as disinfection byproducts or DPBs. So we don't know how prevalent Giardia and Cryptosporidium protozoa are in drinking water, and our means of treatment, if aggressively applied, could cause significant health problems even if the protozoa were removed. What EPA needed was occurrence data for the microbiological species and data on the quantity of disinfection byproducts produced if we treat water with humic acids. Thus was born the Information Collection Rule (ICR).
THE ICR However, EPA got a boost from Congress when language was added to the proposed 1996 SDWA legislation authorizing the agency to require public water suppliers to gather and submit data for determining the need for regulation. EPA was to use the available sampling and analytical methods. This in essence affirmed the ICR. With that data-gathering language intact, the SDWA was passed in August 1996, three months after EPA proposed the initial ICR. The ICR is now in the Code of Federal Regulations [40 CFR 141 Subpart M (140B144)]. EPA had both microbiological and disinfection byproduct methods available for water laboratories to use. The search for Cryptosporidium and other microbial agents, as well as DBPs, could begin. There are three types of data to be collected under the ICR: one set on microbial occurrences, a second set on disinfection byproduct occurrences, and a third on the effect of different types of drinking water treatment processes. The first two data collection efforts began in early 1997 and will run for eighteen months through mid-1998. Treatment study data collection will begin this year and run for twelve months. Water utilities that draw their water from rivers and lakes and serve more than 100,000 people must participate in the ICR. There are approximately 300 such utilities, and if all goes well, data should be available in the year 2000.
SAMPLING AND ANALYSIS At present, the aforementioned 300 public water suppliers are required by the ICR to use approved labs unless they have in-house expertise and the necessary EPA certification. To date, EPA has approved thirty laboratories for these protozoan analyses. A list of the approved labs can be found on EPA's Web site at http://www.epa.gov If ICR data show that microbial contamination is widespread, it is possible that water utilities may want to train their in-house technical staff to run these analyses. However, laboratories not currently certified by EPA but considering offering these microbiological services should be forewarned that Cryptosporidium and Giardia are particularly challenging analyses that require expensive equipment and a highly trained staff. This is not your standard GC/MS analysis. The method specified in the ICR by EPA for Cryptosporidium and Giardia analyses is an immunofluorescent assay. Briefly, 100-1,000 L of water is passed through a 1-µm filter. Particulates are extracted from the filter and centrifuged. The resulting clump of matter is treated with an immunofluorescent stain that attaches itself to any organism. Now comes the good part. An analyst peers through a microscope to see if there are any fluorescently colored particles in the clump or mat of solids. Unfortunately, just being fluorescently colored is not sufficient to identify a Giardia or Cryptosporidium cyst. The analyst must also examine the particles to see which colored specs are the right size and shape to be considered a possible protozoan candidate. Those that are candidates are then examined with another microscope having differential interference contrast optics, which allow the analyst to examine the interior of the cyst. Those cysts that have the right internal morphology (something like seeing the creature inside the human in the movie "Alien") are then classified as hits or confirmed organisms. One sample can take two or more days of an analyst's time to complete. |
|
|
To provide this analytical service, expensive and space-consuming equipment is required. A microscope with immunofluorescence and differential interference contrast optics costs approximately $25,000. If you want to add a still or video camera, you can spend another $10,000. Many microbiological laboratories may have sample preparation and centrifugation equipment already, but chemistry labs that generally provide only simple bacterial (coliform) testing services will need to add these items. |
|
In addition to equipment and supplies, time is another laboratory commodity requiring efficient manage- ment. As you can tell from the above description, a microbiologist must be a patient observer. At a minimum, assuming nothing goes haywire, two and a half days will be needed to complete all the steps from sample preparation to data reporting. Sampling can take from seventy minutes to approximately five hours, depending on the characteristics of the sample water. A highly turbid, high-solids-content water sample will require only ten minutes, as the filters will plug. A clean, clear water sample will require about four hours of filtering; add another hour for preparations and sample handling. Looking through a microscope for Cryptosporidium and Giardia can take six hours. When one analyst and one set of equipment are used, the number of samples that can be processed at one time is limited; pellet processing and viewing are the limiting steps. In a lab where I worked, we were able to conduct these tests only by reducing the process control workload of traditional chemical analyses. For labs wanting to obtain approval to do these protozoan tests, EPA requires that they rack up more than 100 hours of analytical training. One hundred hours of analyzing metals on a graphite furnace atomic absorption spectrophotometer would be more than enough time to experience many of the idiosyncrasies of analysis caused by equipment operation and sample matrices. However, 100 hours of conducting a complex microbiological examination may not be enough experience to do these tests.
TO SIDEBAR: Microbial Agents...Even after conducting approximately seventy-eight hours of analytical work in tandem with an approved laboratory, one municipal lab I am familiar with (not regulated under the ICR but involved in a protozoan study of process wastewater) could not acceptably reproduce a certified lab's data on split samples. Moreover, the lab staff selected for learning these analyses held B.S. degrees in microbiology, were highly skilled in conducting chemical and bacterial analyses, and had taken a three-day hands-on course on Cryptosporidium/Giardia fluorescent immunoassay. The test simply is not easy to perform - looking for Cryptosporidium is like looking for a needle in a haystack. EPA reports that the test is "...highly subjective and dependent on the skill of the analyst..." EPA also reports that mean percent recoveries of 5-44 percent for Giardia cysts and 1-35 percent for Cryptosporidium cysts are common.
TO SIDEBAR: Method 1622I am no expert on the employment market for environmental laboratories, but if the lab where I worked is any indicator, most applicants for lab technician positions were generally trained in biology or microbiology and had a maximum of two years of chemistry. All through the 1970s and 1980s, these biologists were trained to conduct the chemical analyses required by water and waste regulations. Coliform and possibly salmonella tests were the only microbiological tests conducted in many environmental labs, particularly the water and wastewater labs. Now the situation may be altered. Chemists in environmental labs may have to become quasi-biologists.
|
|
TABLE 1. 
